Income Taxation, Transfers and Labor Supply at the Extensive Margin∗ Peter Benczur†, Gabor Katay‡, Aron Kiss§and Oliver M. Racz¶ September 2011
Abstract This paper estimates the effect of income taxation on labor supply at the extensive margin, i.e. the labor force participation. We carry on the estimations on the Hungarian Household Budget Survey (HKF). We extend the methodology of Kimmel and Kniesner (1998) by considering the effect of both taxes and transfers. Nonlabor income contains the (hypothetical) transfer amount someone gets at zero work, while the wage is replaced by the sum of net wages and the amount of lost transfers due to taking up a job (the total change in disposable income if working, CDI). We find that participation probabilities are strongly influenced by transfers and the CDI, particularly for low-income groups. Moreover, the same change in the net wage leads to a much larger change in the CDI for low earners, making them even more responsive to wages and taxation. Our estimates can be readily utilized in tax and transfer reform simulations. Key Words: participation decision, taxation, transfers. JEL Classification: H24, H31, H53, I38, J21 ∗
We thank Gabor Kezdi, Mihaly Szoboszlay, participants at MKE, Atiner 2011, and at seminars at Magyar Nemzeti Bank, Ceska Narodni Banka and the Finance Ministry of France for comments and suggestions. All remaining errors are ours. The views expressed here are of the authors and do not necessarily reflect the official view of MNB. † Magyar Nemzeti Bank and Central European University ‡ Magyar Nemzeti Bank § Magyar Nemzeti Bank ¶ Magyar Nemzeti Bank
1
Introduction
This paper presents a unified parametric approach to estimate the impact of taxes and transfers on the participation decision (the extensive margin of labor supply). The existing literature usually focuses on either taxes or transfers, and mostly adopts a nonparametric, program evaluation framework. Though such approaches are capable of precisely estimating the impact of a particular tax or transfer reform episode, they are unable to serve as ingredients for a simulation of a hypothetical reform’s effects. In our methodology, participation probabilities are determined by the comparison of disposable income in and out of the labor force, consisting of the (often non-observed) amount of transfers and nonlabor income an individual gets if not working and the change in disposable income (CDI) if accepting a job offer (the sum of net wages and lost transfers). This allows a general assessment of the efficiency and effectiveness of government interventions into the labor market, and more importantly, a micro-based prediction of tax and welfare reforms.
85
85 max EU15 (Denmark)
80
80
75 70
71,9
70,1
65 60
75
EU15 74,0 73,9
74,0
70
69,8
68,4 64,7
min EU15 (Italy) 67,2
65
63,1 62,5
60
61,6 59,1
55
55
50
50,8 Turkey
Bulgaria
Croatia
Romania
Malta
Cyprus
Lithuania
Latvia
Estonia
Slovenia
Slovakia
Poland
Czech Republic
45 Hungary
45
50
Figure 1: Participation rates in selected countries, 2007 This issue is particularly relevant for Hungary, where participation rates are among the lowest in the EU (see Figure 1). This has been often identified as a key bottleneck to real convergence (see MNB Convergence Reports 2006, 2008). Motivated by this, Hungary has adopted sweeping changes in its tax and transfer system, by introducing a near flat personal
2
income tax scheme starting in 2010 (an employee tax credit scheme still maintains the zero rate of income below a certain level) and reducing welfare payments starting in 2011. In order to assess the impact of such changes on the participation rate (as done in our related work, Benczur, Katay, Kiss, Reizinger and Szoboszlay, 2011), the incorporation of transfers is important for at least two reasons. As argued for example by the 2008 Convergence Report of MNB, Hungary’s labor participation deficit is mostly due to three special groups, all being highly welfare dependent: women of child-bearing age, the elderly and the low skilled (see Figure 2). Second, for a low-skilled and hence low-wage individual, the change in disposable income from accepting a job offer is heavily influenced by the amount of lost transfers. By following a parametric approach which incorporates both taxes and transfers, our results can thus be utilized to assess the participation effects of a broad tax and welfare system reform.
12
12
10
10
8
8
6
6
4
4
2
2
0
0
-2
-2
-4
-4 Hungary
Poland
women of child-bearing age (15-49)
Slovakia elders (>50)
low-skilled
Czech Republic Total difference
Figure 2: The contribution of certain social groups to the participation rate deficit (percentage points, in 2007)
2
Labor supply and the tax and transfer system
Let us focus first narrowly on taxes. Their impact on labor supply can be summarized in two main tax rate measures. The marginal tax rate measures the tax burden from one extra 3
unit of income. It controls the substitution effect of the intensive margin of labor supply (left panel of Figure 3): leaving household welfare unchanged, a decline in the net wage (a decrease in the marginal tax rate) reduces hours worked from he to h0e . The substitution effect always points in such a direction that an increase in the marginal tax rate leads to a decline in hours worked. The average tax rate, on the other hand, measures the share of the whole labor tax burden from total labor income. A higher average tax rate may encourage labor activity: as long as leisure is a normal good, a higher average tax rate makes consumers poorer, hence they reduce their consumption of leisure. It is also related to the income effect: leaving the local slope of the budget constraint unchanged, a tax increase leads to an inward shift of the budget line, reducing hours worked from h0e to hf (left panel of Figure 3). An increase in taxes may push an individual into a corner solution at zero hours worked (tangency would occur at a negative number of hours), hence also impact the participation decision. Jobs, however, usually have a fixed minimum size (half-time, or in some cases even full-time), which implies that an interior solution at a too low number of hours might also be practically infeasible. In such a situation, a tax increase may discretely push an individual into inactivity. The right panel of Figure 3 depicts such a situation. At the initial net wage w, the individual is supplying he hours of work (say, a full-time job). At a lower net wage w0 , however, the choice between he or zero makes her choose hf = 0. This choice is determined by the average tax rate at her initial gross monthly earnings, since the total amount of tax obligations affects the overall payoff from taking up a job. For this reason, the average tax rate is also labeled as the participation tax rate. (b) Extensive margin
(a) Intensive margin c c=wl+T
c=wl+T
Ui
c
Ui
Ui'
Ui ' c=w'l+T
c=w'l+T T
l
he hf
T
he'
l
he
hf
Figure 3: Impact of a tax change on labor supply Besides influencing nonlabor income (income at zero hours worked), transfers also show characteristics resembling both marginal and average tax rates. Suppose that a certain 4
benefit is means tested with a gradual phaseout. For example, every extra euro earned as wage reduces transfers by 20%. In that case, it is equivalent to a 20% extra marginal tax rate. Once the individual has lost all of this means tested benefit, lost transfers become similar to an average tax rate: the total amount of lost transfers decreases the payoff from work, just like the average tax rate does. If there is again a fixed amount of minimum hours per job, it is only the total amount of transfers lost that matters, similar to a participation tax rate. The overall summary measure in this case is the change in disposable income due to getting a job, which consists of the net wage (for the fixed size of the job) minus the amount of lost transfers.
3
Literature on labor supply responsiveness at the extensive margin
There is a multitude of existing studies which establish that taxes and the welfare system influence the participation decision. There is, however, a notable heterogeneity in terms of implied elasticity measures. This is partly due to the fact that participation regressions are nonlinear, hence the evaluation of the impact of a change in taxes on transfers depend on the characteristics of a given individual or a subgroup. In fact, studies tend to focus on certain narrow social groups, making comparisons even more difficult. Let us briefly review some representative results. Arrufat and Zabalza (1986) do a cross section estimation on the General Household Survey dataset, and find a participation elasticity (the change in the probability of being active in response of a unitary shock in net wages) of η = 1.41 for married women. Dickert et al (1995), using a cross section estimation on Survey of Income and Program Participation (SIPP), find an elasticity of η = 0.2 for single parents. Eissa and Liebman (1996) follow a program evaluation methodology (differences in differences) using the Current Population Survey and the introduction of the Earned Income Tax Credit scheme in the US. They find that single mothers increased their participation rate by 2.8 percentage-points relative to single women without children. Kimmel and Kniesner (1998) adopt a panel estimation on SIPP, and find elasticities of η = [0.6; 2.4; 1.8; 1.1] for single men, single women, wives and husbands respectively. Finally, Aaberge et al (1999) follow a cross section estimation based on the Survey on Household Income and Wealth (Italy), and obtain average elasticities for men and women as η = [0.04; 0.65] respectively. From our point of view, these findings have important shortcomings. Most of them focus on special subgroups, rendering them to limited use for predicting changes in aggregate activity following broad tax and transfer reforms. There is also a substantial heterogeneity
5
in the way after-tax wages are controlled for (if at all), and it is rare that both taxes and transfers are taken simultaneously into account. This is partly due to the fact that recent studies tend to follow a reduced form approach (program evaluation methodology, see Mofitt, 2002 for a review). Though such a framework yields clear results for specific reform episodes, it is not suitable for evaluating the impact of future (hypothetical) scenarios. Meyer and Rosenbaum (2001) is an example of a structural approach, but is not suitable for simulations either: wages are proxied, so the results do not imply a wage elasticity. One major reason for the lack of structural studies that it is not obvious to incorporate all the relevant features of the tax and transfer system into a theory-based framework of labor supply. In the next section, we present an extension of the standard model to incorporate both the marginal and participation tax rate aspect of transfers, but on the expense of constraining the participation decision to a fixed job size.
4 4.1
Theory The underlying theory
The usual approach is to define the reservation wage, which is the threshold for accepting a job offer. Let us start from a standard utility maximization:
s.t.:
(1 − l)1−φ − 1 c1−ψ − 1 +χ max 1−ψ 1−φ c + w (1 − l) = w + T,
where c is consumption, l is labor, w is the wage, and T denotes transfers and other nonlabor income. The total time endowment is normalized to 1, so leisure is 1 − l. The optimality condition can be written as χ (1 − l)−φ = wc−ψ . The reservation wage corresponds to the case where 1 − l∗ = 1. Then c = T , so χ = wres T −ψ defines the reservation wage. The participation decision is then determined by w ≥ wres , or in logs: log w ≥ log χ + ψ log T.
6
Finally, we expand log χi as Zi A0 + εi , where Zi is a vector of observable individual characteristics and εi ∼ N (0, σ 2 ) : log wi − Zi A0 − ψ log Ti ≥ εi . The probability of someone working given a wage offer wi , nonlabor income Ti and individual characteristics Zi is then P =Φ
log wi − Zi A0 − ψ log Ti σ
= Φ γ log wi + Zi α0 − ψ¯ log Ti ,
(1)
yielding the standard structural probit specification. The next step is to add taxes and transfers. One the one hand, we have to modify the wage rate by the effective tax rate (marginal rate, at zero labor income), including taxes, social contributions, and the phaseout of social transfers (if applicable). On the other hand, there are certain transfers which get lost immediately at taking up any job. In such a case, there is a discrete downward jump in T for any nonzero hours worked. One could try to redefine the reservation wage similarly to before, as the level that could still induce an epsilon amount of work. This is, however, not feasible: from Roy’s identity, the welfare gain from a marginal wage increase is the same as the income gain from the extra income due to the higher wage. But there is no such income gain at zero hours worked, so the income equivalent gain is zero, while there is a nonzero income loss due to the drop in T . In other words, the reservation wage is infinite (this can also be established formally by total differentiation). Instead, we redefine the reservation wage by constraining the participation decision to a fixed “job size”˜l – in our empirical specification, it will be a full time job.1 The reservation wage is thus set by the following comparison: • Do not work: then c = T, 1 − l = 1, welfare is
T 1−ψ −1 . 1−ψ
(T −∆T +w˜l) • Work ˜l: then c = T − ∆T + w˜l, 1 − l = 1 − ˜l, welfare is 1−ψ
1−ψ
−1
+χ
(1−˜l)
1−φ
1−φ
−1
.
Introducing the notation W = w˜l − ∆T (the change in disposable income from taking Once working, an individual may decide to work more than ˜l. We assume, however, that it is not known in advance whether there would be opportunities for overtime or performance bonuses, so the activity decision is determined by the base salary. 1
7
up a job, CDI), the comparison becomes: 1−ψ
1−φ 1 − ˜l −1
T 1−ψ − 1 1−φ 1−ψ 1−φ ˜l 1−ψ 1 − −1 1−ψ −1 (T + W ) −1 T − ≥ −χ . 1−ψ 1−ψ 1−φ
(T + W ) −1 +χ 1−ψ
≥
(2)
One can also give a simple graphical representation (see Figure 4): draw the indifference curve going through (C = T, l = 0), find the point of this curve where l = ˜l, and connect this with point (C = T − ∆T, l = 0). Its slope is then the reservation wage: at such a wage level, the individual is just indifferent between not working and getting the full amount of transfers, or working ˜l hours and getting only T − ∆T as transfers.
c c=wl+(T-ΔT)
Ui
T T-ΔT
l
l*
Figure 4: The reservation wage when there is a discrete drop in transfers To derive a formal expression for the probabilty of being active (the analogue of (1)), let us linearize the left hand side of (2): (T + W )1−ψ − 1 T 1−ψ − 1 − ≈ W T −ψ , 1−ψ 1−ψ
8
so the comparison becomes
W T −ψ ≥ χ
1−φ 1 − 1 − ˜l 1−φ {z
|
Q
= χQ. }
The individual works if log W − ψ log T − log χ − log Q ≥ ε, yielding again a structural probit of the form P = Φ γ log Wi + Zi α0 − ψ¯ log Ti .
(3)
Let us compare the two structural probit equations (1) and (3). In the latter, Zi should contain the constant; but it should be there anyway. Second, Wi in (3) is the change in disposable income due to getting a full time job: Wi = wi ˜l − ∆Ti , as opposed to the net wage wi . Third, Ti is the hypothetical amount of transfers one gets (or would get) at zero hours worked. From a practical point of view, T is not directly observable for the employed, since they get T − ∆T ; while ∆T is not observed for the inactive, since they get T . Using individual characteristics and the welfare system’s details (for every given year), however, one can back up T and ∆T . It essentially requires a microsimulation tool. For those who work, we determine T based on their characteristics and welfare regulations for the given year, and then obtain ∆T = T − Tobs . For those who do not work, we determine ∆T by again applying welfare rules, while T = Tobs .
5
Econometric issues
Here we closely follow Kimmel and Knieser (1998), up to a certain point. We want to estimate a structural probit equation: P (employed/active) = Φ γ log Wi + Zi α0 − ψ¯ log Ti , were Wi = wi ˜l − ∆Ti . Here the vector Zi contains individual characteristics which shift the labor supply of an individual. As usual in the literature on participation, there is a missing data issue: the wage is unavailable for those who do not work. The solution is to use a
9
predicted W for the inactives – run log Wi = Xi β 0 + µi ˆ = Xi βb for the unemployed. Here the vector for the employed, and use the predicted wage W Xi contains individual characteristics which are relevant for defining an individual’s wage. Note that the two vectors Xi and Zi may overlap, but there can be elements in each of them which are excluded from the other set. This regression, however, is run on a nonrandom sample, since the employment and the W error terms might be correlated. The solution is thus to adopt a Heckman-type correction, yielding a three step procedure. In variant A, we thus adopt the following procedure: 1. Run a reduced form probit 0 0 P (employed) = Φ (Xi βRF + Zi αRF − ψRF log Ti ) .
2. Use the inverse Mills ratio λ (x) =
φ(x) Φ(x)
as a correction in the log CDI regression:
0 0 log Wi = Xi β 0 + δλ Xi βbRF + Zi α bRF − ψbRF log Ti + µi . \i = Xi βb0 in the structural probit equation 3. Use the predicted log CDI logW
0 ¯ \ P (employed/active) = Φ γ log W i + Zi α − ψ log Ti . Notice that here X ⊇ Z, since there is practically no observable characteristics which would not be related to transfer measures, which are there in log W . In variant B, we slightly modify the previous procedure: 1. Run a reduced form probit 0 0 − ψRF log Ti ) . P (employed) = Φ (Xi βRF + Zi αRF
2. Use the inverse Mills ratio λi (x) = monthly income) regression:
φ(x) Φ(x)
as a correction in the wage (more precisely:
0 0 + Zi α bRF − ψbRF log Ti + µi . log wi = Xi β 0 + δλ Xi βbRF
10
[ 3. Use the predicted log wage log wi = Xi βb0 , exponentiate, subtract ∆Ti and take logs again to obtain the predicted log CDI for the structural probit equation \ P (employed/active) = Φ γ log W i + Zi α0 − ψ¯ log Ti . Three remarks are in order. The first is regarding endogeneity. In the structural probit, log W can be endogenous, since the wage error term can be correlated with the participation decision error term. Notice, however, that we are in fact running an IV-probit in step 3, so this endogeneity issue is automatically handled (as long as there are variables in Xi which are excluded form Zi , an issue we address in the data section). The second issue is whether the selection correction is identified only through a functional form assumption. This is indeed the case when X ⊇ Z, since the inverse Mills ratio is then just a nonlinear reshuffling of the right hand side variables in the wage equation (variant A). On the other hand, the inverse Mills ratio does contain additional variation if X # Z, which is the case in Variant B. This means that we are free from this criticism in Variant B, but it applies in Variant A. In that case, however, there is no alternative: if a variable impacts the participation equation directly, it is also likely to impact the CDI (log W ) at least through the change in transfers term ∆T . Finally, there is a technical issue in Variant B (yet to be resolved). In step 2, we are obtaining a consistent estimate for log w. When we move to log (w − ∆T ), one has to add a correction term due to the nonlinearity of the log function, which (under some normality conditions) should contain some cov(w, ∆T ) terms. Though the conditions under which one can derive such an explicit correction term may put strong restrictions on the joint distribution of w and ∆T , there is a similar implicit restriction in Variant A: it assumes that individual errors are multiplicative for the CDI as opposed to the log wage, which is a also restriction on the joint distribution of the wage and the change in transfers. As of now, we follow Variant A, but later we would switch to Variant B (and discuss Variant A only as a robustness check). Since our “wage”measure in the structural estimation is the CDI, the calculation of regular wage elasticities requires one more step. The structural probit gives us a log CDI coefficient γ. Since the probit is a nonlinear function, one has to evaluate it at a certain vector Z and log T to obtain the marginal impact of a percentage change in the CDI. Even then, however, it is still the impact of a change in W , not w. To obtain the impact of the wage itself, note that ∂ log elog w − ∆T elog w w ∂ log (w − ∆T ) = = log w = , ∂ log w ∂ log w e − ∆T w − ∆T 11
so
∂Φ ∂Φ ∂ log W ∂Φ w = = . ∂ log w ∂ log W ∂ log w ∂ log W w − ∆T
(4)
Notice that the marginal effect of log W gets magnified if w − ∆T w – which is the case for transfer-dependent people (low skill, around retirement, etc.)
6
Data
We use data from the Hungarian Household Budget Survey (HKF), years 1998-2008. This is a rotating panel database with a one-third renewing part every year, but we only use it as a pooled cross-section. The dataset contains detailed income and consumption measures of broadly 25,000 individuals per year. The key challenge is to define the counterfactual transfers: First, how much would someone who is currently working get if gets laid off? Second, how much would someone who is currently inactive lose if takes up a full time job? Calculating these measures requires the detailed coding of the full transfer system, basically a microsimulation model. With one exception, the database contained all the relevant information to deduct the counterfactual transfer entitlements or losses of each individual. The exception was the work history of individuals, on which certain transfers depend (for example, eligibility to the more generous maternity support schedule GYES). To resolve this issue, we used a predicted value based on the Labor Force Survey database (a conditional expectation based on observable characteristics). The main left hand side variable was labor force participation, though we also ran the same estimations with employment. The right hand side measures form two major groups: labor-supply shifters (Zi ) and wage equation controls (Xi ). Following MaCurdy (1985, 1987) and Kimmel and Kniesner (1998), labor-supply shifters contain personal and family characteristics, while the vector Xit includes variables which determine the market wage, mostly interactions of individual characteristics with schooling. In particular, the first group consists of the following variables: non-labor income, education, household head, mother with infant (< 3 years old), attending full-time education, household size (number of persons), pensioner, family status (husband, wife, child, single, divorced,. . . ), and year dummies. The second group contains age, age squared, education, gender, county, year dummies, and a full set of interactions with education. In Variant A, however, both groups are entering the CDI equation.
12
7
Results
This section reports and discusses our empirical results. Though we report numbers both for participation and employment as left hand side variables, we focus on the participation margin. Table 1 displays our baseline results, following the econometric methodology of Variant A. Panel A reports the estimates for the structural probit equation (3). Most point estimates have the expected sign and are significant. A higher CDI increases the probability of being active, while nonlabor income has the opposite effect (both are in logs). Education seems to have a mixed effect: the coefficient of the (at most) elementary school dummy is negative, vocational is positive, secondary is negative but insignificant, while tertiary is significantly negative. Being a household head or having a larger family increases activity, while being a mother with infant, full-time student or pensioner decreases it. With some exceptions (mostly for education), the results are quite similar when the left hand side variable is employment. (A) Estimation results
net wage - transfer non labour income elementary school vocational secondary education tertiary education head of household mother with infant full-time student household size pensioner
participation (1) coef. std. err. 0.895 0.018 -1.041 0.023 -0.113 0.032 0.077 0.034 -0.044 0.036 -0.113 0.042 0.304 0.023 -0.406 0.017 -2.336 0.031 0.113 0.005 -2.139 0.027
employment (2) coef. std. err. 0.842 0.017 -0.756 0.022 0.053 0.031 0.259 0.032 0.237 0.034 0.227 0.039 0.302 0.019 -0.224 0.015 -2.078 0.035 0.085 0.005 -1.758 0.026
(B) Conditional marginal effects net wage - transfer non labour income net wage transfer
dy/dx 0.248 -0.288 0.349 -0.162
std. err. 0.002 0.003 0.003 0.001
dy/dx 0.295 -0.265 0.416 -0.177
std. err. 0.004 0.006 0.005 0.003
Table 1: Main results Since the probit function is highly nonlinear, the point estimates in Panel A are not indicative about the conditional marginal effect of variables of interest on activity. Panel 13
B displays these numbers, evaluated at the sample means. Numbers here are already semielasticities: a 10% increase in the CDI leads to a 2.48% increase in the probability of being active. As explained by equation (4), the same increase in the net wage (as opposed to the net wage minus transfers) leads to a potentially larger effect. The difference is quite substantial at the sample mean, as the effect is almost 50% higher. The opposite happens with nonlabor income: transfers are only part of them, so a 10% change in transfers implies a smaller increase in nonlabor income. Next we look at the conditional marginal effects by subgroups to see how much they differ from each other. Table 2 presents two variants, a full and a restricted sample estimate. The full sample means that all observations are included (as in table 1), but the marginal effects are evaluated at a subgroup-specific mean. The restricted sample means that the entire estimation procedure is carried out only on the subsample at hand, so even the structural probit estimates can be different. participation (1) full sample
elementary school or less
secondary education
tertiary education
without self-employed
prime-age (25-54)
net wage - transfer non labour income net wage transfer net wage - transfer non labour income net wage transfer net wage - transfer non labour income net wage transfer net wage - transfer non labour income net wage transfer net wage - transfer non labour income net wage transfer
dy/dx 0.318 -0.370 0.488 -0.212 0.185 -0.215 0.270 -0.138 0.090 -0.105 0.115 -0.053 0.271 -0.315 0.388 -0.184 0.067 -0.078 0.099 -0.059
employment (2)
restricted sample
std. err. 0.009 0.011 0.013 0.006 0.002 0.003 0.002 0.001 0.002 0.002 0.003 0.001 0.003 0.004 0.004 0.002 0.001 0.001 0.001 0.001
dy/dx 0.229 -0.350 0.393 -0.204 0.194 -0.211 0.274 -0.131 0.092 -0.093 0.117 -0.050 0.271 -0.376 0.376 -0.184 0.068 -0.074 0.099 -0.057
std. err. 0.007 0.012 0.012 0.006 0.005 0.007 0.007 0.004 0.003 0.003 0.004 0.002 0.006 0.009 0.008 0.004 0.001 0.001 0.001 0.001
full sample dy/dx 0.240 -0.215 0.368 -0.153 0.248 -0.222 0.362 -0.169 0.123 -0.110 0.156 -0.063 0.309 -0.278 0.443 -0.192 0.128 -0.115 0.189 -0.100
std. err. 0.008 0.009 0.012 0.005 0.003 0.005 0.004 0.002 0.002 0.003 0.003 0.001 0.004 0.007 0.006 0.003 0.001 0.002 0.001 0.001
restricted sample dy/dx 0.223 -0.233 0.383 -0.187 0.209 -0.138 0.295 -0.120 0.126 -0.112 0.160 -0.064 0.275 -0.263 0.381 -0.162 0.120 -0.087 0.175 -0.085
std. err. 0.008 0.013 0.014 0.007 0.008 0.011 0.011 0.006 0.004 0.005 0.005 0.002 0.007 0.011 0.010 0.005 0.002 0.004 0.003 0.003
Table 2: Conditional marginal effects by subgroups Notice that the net wage (or even the CDI) elasticity of activity is highly different across the three educational groups even in the full sample estimation case, when the only reason is a different conditional mean of the subgroups. The exclusion of the self-employed barely effects the results, while all elasticities become much smaller in the prime-age subgroup. The restricted sample delivers similar results, though for the group with at most elementary 14
school education the restricted sample estimates are slightly lower. Table 3 further explores the prime-age sample, checking whether education status also matters there. The low overall elasticity of this age group which was shown in Table 2 splits into a sizable elasticity for the elementary school or less group (a group which is also highly welfare dependent) and a small but still significant number for the other two groups. The restricted samples yield similar though smaller differences. participation (1) full sample
elementary school or less
secondary education
tertiary education
net wage - transfer non labour income net wage transfer net wage - transfer non labour income net wage transfer net wage - transfer non labour income net wage transfer
dy/dx 0.207 -0.241 0.358 -0.218 0.067 -0.078 0.102 -0.063 0.029 -0.034 0.038 -0.022
employment (2)
restricted sample
std. err. 0.002 0.003 0.003 0.002 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
dy/dx 0.113 -0.142 0.192 -0.119 0.055 -0.057 0.084 -0.048 0.032 -0.031 0.042 -0.021
std. err. 0.009 0.013 0.014 0.010 0.002 0.003 0.003 0.002 0.003 0.004 0.004 0.002
full sample dy/dx 0.286 -0.257 0.493 -0.279 0.130 -0.116 0.198 -0.109 0.054 -0.049 0.071 -0.035
std. err. 0.003 0.005 0.006 0.004 0.001 0.002 0.002 0.001 0.001 0.001 0.002 0.001
restricted sample dy/dx 0.212 -0.158 0.360 -0.193 0.080 -0.023 0.121 -0.049 0.037 -0.022 0.048 -0.019
std. err. 0.009 0.016 0.015 0.011 0.006 0.009 0.009 0.006 0.007 0.009 0.009 0.005
Table 3: Conditional marginal effects by subgroups, prime age subsample Finally, Table 4 displays the conditional marginal effects for the two remaining main welfare dependent social groups, the elderly and women of child-bearing age. The group of age above 50 exhibits a very substantial elasticity – this partly explains the large gap between the elasticity of the entire population and the prime-age group. This finding is quite important, as it shows that taxes and transfers have a strong impact on activity around retirement age, and that the tax and social insurance system can contribute to the large activity gap of the elderly in Hungary. Maybe surprisingly, women at child-bearing age show a modest wage elasticity, though they are still more responsive than the overall prime-age group. This is also true about the impact of transfers. In summary, we have found that wages, taxes and transfers have a large impact on the participation decision, particularly for elders, the low-skilled, and to some degree, women at child-bearing age. Moreover, these differences can be largely explained by different group characteristics, leading to different conditional marginal effects of the same structural probit estimates, and also to a different multiplication of a net wage change into the change in the CDI.
15
elders (>50)
women at child-bearing age (25-49)
net wage - transfer non labour income net wage transfer net wage - transfer non labour income net wage transfer
participation (1) dy/dx std. err. 0.357 0.007 -0.415 0.009 0.458 0.009 -0.147 0.003 0.114 0.001 -0.133 0.002 0.184 0.002 -0.115 0.001
employment (2) dy/dx std. err. 0.330 0.007 -0.296 0.009 0.423 0.009 -0.126 0.003 0.163 0.001 -0.146 0.002 0.262 0.002 -0.150 0.001
Table 4: Conditional marginal effects for elders and women at child-bearing age
8
Conclusion
This paper presents a first (at least to our knowledge) structural form estimation of labor supply at the extensive margin that simultaneously takes into account taxes and transfers. We show that one has to modify the net wage by deducting the amount of lost transfers to get the measure which determines the participation decision (the change in disposable income). This implies, however, that the same change in the net wage leads to a very different change in the CDI if lost transfers are a different share of the net wage. We find that a single equation can already explain a large heterogeneity of individual responsiveness to taxes and transfers: there are large differences among subgroups, driven partly by a composition effect, and partly by a different share of lost transfers in the CDI. These highly responsive subgroups are exactly the ones who are mostly responsible for Hungary’s low participation rate (low-skilled, women at child-bearing age, elders), implying that a reform of the tax and transfer system can be a powerful tool to boost employment. Our results in fact directly lend themselves to reform simulations. In related work (Benczur, Katay and Kiss, 2011), we build a model where labor supply is determined through a detailed microsimulation model, based on our results here for the extensive margin, and a combination of Bakos, Benczur and Benedek (2008) and Kiss and Mosberger (2011) for the intensive margin. We briefly sketch the mechanics of our simulation tool. Concentrating only on aggregate activity, the main step is to calculate the change in labor supply (at the extensive margin) for a given (hypothetical) wage, and tax and transfer system. First we obtain the pre- and post-reform aftertax wage income of everyone, using either observed or predicted wages. Then we calculate the pre- and post-reform hypothetical 16
“zero hours worked”transfer level for everyone, and construct the (log of the) CDI (log W ) before and after the reform. Equipped with these, we form
b¯ log T Φ γ b log Wi + Zi α b −ψ i 0
before and after the reform. The change in its value is the change in the probability of individual i being employed. Finally, we add up the probabilities in the sample (weighted) to get an estimate for the change in the aggregate activity rate. For a full “micro-based macro model”, we add a similar exercise on the intensive margin, and then put the simulation results from both the intensive and extensive margins in a simple macro model, in which an estimated production function determines labor and capital demand, and there is an almost perfectly elastic capital supply function (motivated by a small open economy assumption). For a given reform, the microsimulation unit gives us the shift in the labor supply curve. Profit maximization and market clearing then, through a simple still general equilibrium mechanism, tells us the decline in the gross wage and the increase in total (effective) labor consistent with supply and demand. In an earlier version (Benczur and Katay, 2010), we applied such a framework (using the same intensive but an ad hoc extensive margin treatment) to evaluate the 2010 Hungarian tax reform (see MNB Inflation Report, December 2010). With a fully fledged extensive margin model and a detailed inclusion of transfers, we are able to evaluate at depth the 2011 tax and transfer reforms as well (Benczur, Katay, Kiss, Reizinger and Szoboszlay, 2011).
References Aaberge, Rolf., Ugo Colombino and Steinar Strom (1999): “An Empirical Analysis of Joint Household Decisions, with Taxes and Quantity Constraints”, Journal of Applied Econometrics, Vol. 14, No. 4, pp. 403-422 Arrufat, Jose L. and Antonio Zabalza (1986): “Female Labor Supply with Taxation, Random Preferences, and Optimization Errors”, Econometrica, Vol. 54, No. 1, pp. 47-63 Bakos, Peter, Peter Benczur and Dora Benedek (2008): “The Elasticity of Taxable Income: Estimates and Flat Tax Predictions using the Hungarian Tax Changes in 2005”, MNB Working Papers, 2008/7 Benczur, Peter and Gabor Katay (2010): “Assessing the Impact of Tax Reforms on Hungary with a General Equilibrium Model”(in Hungarian), mimeo, Magyar Nemzeti Bank Benczur, Peter, Gabor Katay and Aron Kiss (2011): “Assessing the Impact of Tax
17
and Transfer Reforms with a General Equilibrium Microsimulation Model”, mimeo, Magyar Nemzeti Bank Benczur, Peter, Gabor Katay, Aron Kiss, Balazs Reizinger and Mihaly Szoboszlay (2011): “Analyzing the Impact of Hungarian Tax and Transfer Reforms with a Microsimulation Model”, forthcoming in MNB Bulletin 2011/3. Dickert, Stacy, Scott Houser and John K. Scholz (1995): “The Earned Income Tax Credit and Transfer Programs: A Study of Labor Market and Program Participation”, in James M. Poterba (ed.), Tax Policy and the Economy, Vol. 9, MIT Press, pp. 1-50 Eissa, Nada and Jeffrey B. Liebman (1996): “Labor Supply Response to the Earned Income Tax Credit”, Quarterly Journal of Economics, Vol. 61, No. 2, pp. 605-637 Kimmel, Jean and Thomas J. Kniesner (1998): “New Evidence on Labor Supply: Employment versus Hours Elasticities by Sex and Marital Status”, Journal of Monetary Economics, Vol. 42., pp. 289-301 Kiss, Aron and Palma Mosberger (2011): “The Elasticity of Taxable Income of High Income Earners: Evidence from Hungary”, mimeo, Magyar Nemzeti Bank MaCurdy, Thomas E. (1985): “Interpreting Empirical Models of Labor Supply in an Intertemporal Framework with Uncertainty”, in Heckman, James J. and Burton Singer (eds.), Longitudinal Analysis of Labor Market Data, Cambridge University Press, pp. 111-155 MaCurdy, Thomas E. (1987): “A Framework for Relating Microeconomic and Macroeconomic Evidence of Intertemporal Substitution”, in Truman F. Bewley, (ed.), Advances in Econometrics, Fifth World Congress of the Econometric Society, Vol. 2, Cambridge University Press, pp. 149-176 Magyar Nemzeti Bank (2006, 2008): “Analysis of the Convergence Process”, years 2006 and 2008 Magyar Nemzeti Bank (2010): “Quarterly Report on Inflation Report”, November 2010, pp. 46-50 Meyer, Bruce D. and Dan T. Rosenbaum (2001): “Welfare, The Earned Income Tax Credit, and the Labor Supply of Single Mothers”, Quarterly Journal of Economics, Vol. 116, No. 3, pp. 1063-1114 Mofitt, Robert (2002): “Welfare Programs and Labor Supply”, in Alan J. Auerbach and Martin Feldstein (eds.), Handbook of Public Economics, Vol. 4., Elsevier, pp. 2393-2430
18
